Introduction of biologic agents, including anti-cytokine and B cell-depleting antibodies, into the therapeutic arsenal has resulted in considerable progress in the clinical management of rheumatoid arthritis (RA), a debilitating autoimmune disease affecting the synovial joints. However, a significant proportion of patients with RA do not respond adequately to biologic agents, and for those, novel therapeutic approaches are needed. Data from clinical trials indicate that many RA patients with severe treatment-refractory disease respond well to autologous bone marrow transplantation (ABMT). The clinical success of ABMT is primarily attributed to hematopoietic stem cell-mediated restoration of immune homeostasis and self tolerance through induction of immunosuppressive regulatory T cells (Tregs). We hypothesize that innate immune cells such as myeloid- derived suppressor cells (MDSCs) that are likely present in, and released from, the patients' bone marrow (BM) to the circulation, contribute to the improvement of RA symptoms in ABMT recipients. Our hypothesis is based on the observations that MDSCs are present in mice with proteoglycan (PG)-induced arthritis (PGIA), an inducible autoimmune animal model of RA, and that mice with PGIA also show significant improvement upon ABMT. In preliminary studies we have identified innate immune cells in the synovial fluid, BM, and spleens of arthritic mice, which exhibit an immature myeloid cell phenotype and profound suppressor activity toward activated T lymphocytes in vitro, partly by inhibiting the maturation of dendritic cells into potent antigen presenting cells. MDSCs that are able to antagonize T cell activation accumulate in tumor-bearing individuals and suppress anti-tumor immune responses. Although MDSCs with a potential to suppress autoimmunity likely exist in RA and animal models of RA, the presence of such cells in individuals with autoimmune arthritis has not been reported to date. The studies described in Aim 1 will identify MDSC subsets that accumulate at distinct anatomical sites in mice with PGIA, and analyze the trafficking patterns, suppressor activity of MDSCs, and the mechanisms of suppression, upon the induction and progression of arthritis. In Aim 2, we will test the potential of MDCSs to prevent or suppress disease by transferring these cells from arthritic mice to syngeneic animals before or after the development of PGIA. Although conceptually similar to ABMT, this treatment will be reduced to the adoptive transfer of MDSCs, i.e., purified populations of innate immune cells that have the potential to subvert the activation of autoreactive T cells in the recipient. As a translational approach (Aim 3), we propose to identify and functionally characterize MDSCs in the blood and synovial fluid of RA patients, and determine whether therapeutically useful quantities of MDSCs can be generated through an in vitro cell enrichment strategy. The results of our studies should provide the foundation and experimental framework of future investigations into MDSC function in autoimmune arthritis, and may also open a new direction toward the development of autologous MDCS transfer-based treatment strategies in RA.